Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Xbench 1.1 is now available. Optimized for the PowerMac G5.

The new version gives my Quicksilver 867 (single) 512MB

4 less Gigaflops on Altivec.
AAnd this vecLib FFT (I think thats it) I get a 20 on it.... not cool.
 
Its weird... I thought I posted in this thread... hmmm well my post from the other thread really says what I want:

Originally posted by MrMacman
Is it me or do you guys think there should be 2 benchmarks:

Optomized for the G5 And
Not Optimized for the G5.

Look, right now most products are *not* optomized for the G5, so optomizing benchmarking is almost pointless.

Its giving the advantage to the G5 almost un-fairly.

You must expect they don't optomize to have a fair benchmark.
Click to expand...

Anyone else think my idea is correct?

Your gonna have Photoshop all optomized soon enough, now xBench is optomized, you know what, when we have a non-optomized app your not gonna get the preformance your thinking of.
 
Originally posted by MrMacman
Its weird... I thought I posted in this thread... hmmm well my post from the other thread really says what I want:



Anyone else think my idea is correct?

Your gonna have Photoshop all optomized soon enough, now xBench is optomized, you know what, when we have a non-optomized app your not gonna get the preformance your thinking of.
Click to expand...

Well, You're always going to have benchmarks different for different apps. I think that having two versions of xBench is a fine idea, but just having one that is optimized for the G5 is better than having just one that isn't. Apps are going the 64 bit route (if needed) so I think that the new xBench is just fine.

P-Worm
 
Originally posted by MrMacman
Anyone else think my idea is correct?

Your gonna have Photoshop all optomized soon enough, now xBench is optomized, you know what, when we have a non-optomized app your not gonna get the preformance your thinking of.
Click to expand...

Well the thing is benchmarks are optimized for other processors, so to have a truly fair competition it needs to be optimized for the G5 too. Xbench 1.0 was optimized for the G4 processor, and when dealing with cross-platform benchmarks they are optimized for the Pentium and Athlon processors
 
The bench results for PowerMac G5 1.6 ghzfrom SpyMac

Guys SteveG56 posted these results on SpyMac forums. The results are as listed below:

Longer is better — scores greater than 200 will be truncated, and have "—>" inside the bar

Name
Score
Detail

Results
126.13

System Info

Xbench Version
1.1

System Version
10.2.7

Physical RAM
1280 MB

Model
PowerMac7,2

Processor
PowerPC 970 @ 1.60 GHz

L1 Cache
64K (instruction), 32K (data)

L2 Cache
512K @ 1600 MHz

Bus Frequency
800 MHz

Video Card
GeForce FX 5200

Drive Type
ST380013AS

CPU Test
91.85

GCD Recursion
72.97
2.85 Mops/sec

Floating Point Basic
213.26—>
758.44 Mflop/sec

AltiVec Basic
78.43
1.14 Gflop/sec

vecLib FFT
51.25
789.76 Mflop/sec

Floating Point Library
264.65—>
10.59 Mops/sec

Thread Test
80.68

Computation
59.27
477.48 Kops/sec, 4 threads

Lock Contention
126.33
1.59 Mlocks/sec, 4 threads

Memory Test
251.00—>

System
199.11

Allocate
327.59—>
110.45 Kalloc/sec

Fill
155.75
1239.80 MB/sec

Copy
178.75
893.76 MB/sec

Stream
339.49—>

Copy
335.86—>
1369.69 MB/sec [G5]

Scale
319.23—>
1326.10 MB/sec [G5]

Add
347.79—>
1467.62 MB/sec [G5]

Triad
357.52—>
1471.49 MB/sec [G5]

Quartz Graphics Test
156.39

Line
174.55
4.44 Klines/sec [50% alpha]

Rectangle
174.20
12.26 Krects/sec [50% alpha]

Circle
175.11
4.04 Kcircles/sec [50% alpha]

Bezier
148.88
1.62 Kbeziers/sec [50% alpha]

Text
123.84
2.02 Kchars/sec

OpenGL Graphics Test
188.62

Spinning Squares
188.62
132.00 frames/sec

User Interface Test
141.20

Elements
141.20
48.04 refresh/sec

Disk Test
105.75

Sequential
114.68

Uncached Write
146.61
58.36 MB/sec [4K blocks]

Uncached Write
138.07
53.89 MB/sec [256K blocks]

Uncached Read
75.75
11.99 MB/sec [4K blocks]

Uncached Read
131.30
53.05 MB/sec [256K blocks]

Random
98.12

Uncached Write
86.38
1.24 MB/sec [4K blocks]

Uncached Write
102.33
23.08 MB/sec [256K blocks]

Uncached Read
95.15
0.63 MB/sec [4K blocks]

Uncached Read
112.26
23.10 MB/sec [256K blocks]

I don't know how the G5 compares to the Dual 1.25 or Dual 1.42, please post your result in comparison.
 
Re: The bench results for PowerMac G5 1.6 ghzfrom SpyMac

Results 122.37
System Info
Xbench Version 1.1
System Version 10.2.6
Physical RAM 1024 MB
Model PowerMac3,6
Processor PowerPC G4x2 @ 1.00 GHz
Version 7455 (Apollo) v2.1
L1 Cache 32K (instruction), 32K (data)
L2 Cache 256K @ 1000 MHz
L3 Cache 1024K @ 250 MHz
Bus Frequency 167 MHz
Video Card GeForce4 Ti 4600
Drive Type IBM-IC35L080AVVA07-0
CPU Test 124.84
GCD Recursion 124.15 4.85 Mops/sec
Floating Point Basic 122.76 436.57 Mflop/sec
AltiVec Basic 125.86 1.83 Gflop/sec
vecLib FFT 125.90 1.94 Gflop/sec
Floating Point Library 125.61 5.03 Mops/sec
Thread Test 117.85
Computation 119.12 959.65 Kops/sec, 4 threads
Lock Contention 116.61 1.46 Mlocks/sec, 4 threads
Memory Test 144.54
System 132.37
Allocate 124.69 42.04 Kalloc/sec
Fill 208.02 1655.87 MB/sec
Copy 101.67 508.35 MB/sec
Stream 159.16
Copy 156.28 637.34 MB/sec [altivec]
Scale 158.85 659.86 MB/sec [altivec]
Add 165.90 700.09 MB/sec [altivec]
Triad 156.00 642.08 MB/sec [altivec]
Quartz Graphics Test 121.67
Line 128.69 3.28 Klines/sec [50% alpha]
Rectangle 122.91 8.65 Krects/sec [50% alpha]
Circle 134.52 3.10 Kcircles/sec [50% alpha]
Bezier 122.85 1.33 Kbeziers/sec [50% alpha]
Text 104.02 1.70 Kchars/sec
OpenGL Graphics Test 116.77
Spinning Squares 116.77 81.72 frames/sec
User Interface Test 95.16
Elements 95.16 32.38 refresh/sec
Disk Test 94.19
Sequential 95.77
Uncached Write 88.05 35.05 MB/sec [4K blocks]
Uncached Write 83.07 32.42 MB/sec [256K blocks]
Uncached Read 155.35 24.59 MB/sec [4K blocks]
Uncached Read 83.80 33.86 MB/sec [256K blocks]
Random 92.66
Uncached Write 96.10 1.37 MB/sec [4K blocks]
Uncached Write 90.39 20.39 MB/sec [256K blocks]
Uncached Read 95.67 0.63 MB/sec [4K blocks]
Uncached Read 88.93 18.30 MB/sec [256K blocks]
 
This xbench version has to be wrong. My Altivec basic went from 4+ GFlops to 1 and some change. Every test is down across the board. I just don't believe that a 1.6 G5 is only going to have 1.75 GFlops Altivec Basic either.

This update didn't work... I imagine there will be a 1.1.1 soon.
 
I also felt that the Xbench results were a bit screwed up. I hope that the newer version of Xbench is out before I receive my PowerMac G5, so that I could do some real benchmark tests.
 
Re: spelling

Originally posted by phc
Learn how to spell!

O-P-T-I-M-I-Z-E
Click to expand...

:lol: thank you, I was about to post that myself. Are you on OS X? Turn on spelling as you type in this text field (the "Your Reply:" one).

I don't like this new Xbench, since my quicksilver has a higher CPU rating than the G5. However, my memory test was much, much lower :lol:

There's a serious problem with that Veclib FFT. Some G4's are getting realllllly low (I got a 20), some are getting quite high (see above Dual 1Ghz posted by Chaszmyr), and the G5 is at 50!
 
Re: Re: spelling

Originally posted by bobindashadows
There's a serious problem with that Veclib FFT. Some G4's are getting realllllly low (I got a 20), some are getting quite high (see above Dual 1Ghz posted by Chaszmyr), and the G5 is at 50!
Click to expand...

The first time i ran the new xbench my scores were VERY low, but they were back to normal subsequent times
 
obviously, no xbench

I've been hearing about extremely erratic results with this program for a while, so I don't know why people keep using it. If you test the same system twice, I've heard you can get very different results. I just don't think it's well-written. we should stick to cinebench, PShop, UT2003, Quake3, that type of professionally coded cross-platform benchmark that doesn't have radically different results every time you run it, or get insanely slower with an update.
 
Re: Re: The bench results for PowerMac G5 1.6 ghzfrom SpyMac

Originally posted by Chaszmyr
Code: 
Results	122.37	
	System Info		
		Xbench Version		1.1
		System Version		10.2.6
		Physical RAM		1024 MB
		Model		PowerMac3,6
		[B]Processor		PowerPC G4x2 @ 1.00 GHz[/B]
			Version		7455 (Apollo) v2.1
			L1 Cache		32K (instruction), 32K (data)
			L2 Cache		256K @ 1000 MHz
			L3 Cache		1024K @ 250 MHz
			Bus Frequency		167 MHz
		Video Card		GeForce4 Ti 4600
		Drive Type		IBM-IC35L080AVVA07-0
	CPU Test	124.84	
		GCD Recursion	124.15	4.85 Mops/sec
		Floating Point Basic	122.76	436.57 Mflop/sec
		AltiVec Basic	125.86	1.83 Gflop/sec
		vecLib FFT	125.90	1.94 Gflop/sec
		Floating Point Library	125.61	5.03 Mops/sec
	Thread Test	117.85	
		Computation	119.12	959.65 Kops/sec, 4 threads
		Lock Contention	116.61	1.46 Mlocks/sec, 4 threads
	Memory Test	144.54	
		System	132.37	
			Allocate	124.69	42.04 Kalloc/sec
			Fill	208.02	1655.87 MB/sec
			Copy	101.67	508.35 MB/sec
		Stream	159.16	
			Copy	156.28	637.34 MB/sec [altivec]
			Scale	158.85	659.86 MB/sec [altivec]
			Add	165.90	700.09 MB/sec [altivec]
			Triad	156.00	642.08 MB/sec [altivec]
	Quartz Graphics Test	121.67	
		Line	128.69	3.28 Klines/sec [50% alpha]
		Rectangle	122.91	8.65 Krects/sec [50% alpha]
		Circle	134.52	3.10 Kcircles/sec [50% alpha]
		Bezier	122.85	1.33 Kbeziers/sec [50% alpha]
		Text	104.02	1.70 Kchars/sec
	OpenGL Graphics Test	116.77	
		Spinning Squares	116.77	81.72 frames/sec
	User Interface Test	95.16	
		Elements	95.16	32.38 refresh/sec
	Disk Test	94.19	
		Sequential	95.77	
			Uncached Write	88.05	35.05 MB/sec [4K blocks]
			Uncached Write	83.07	32.42 MB/sec [256K blocks]
			Uncached Read	155.35	24.59 MB/sec [4K blocks]
			Uncached Read	83.80	33.86 MB/sec [256K blocks]
		Random	92.66	
			Uncached Write	96.10	1.37 MB/sec [4K blocks]
			Uncached Write	90.39	20.39 MB/sec [256K blocks]
			Uncached Read	95.67	0.63 MB/sec [4K blocks]
			Uncached Read	88.93	18.30 MB/sec [256K blocks]
Click to expand...
Now I can read it.;)
 
If (insert your app/task here)can be optimised to run on G5 much faster than it ever could on G4, that'd be the real indicator of performance advantage. For me at least.

Personally, I think that benchmarks are rarely anything more than marketing tool for hardware vendors.
 
I have got some very odd scores with this program as well, i found that just moving the Xbench app from a folder to my desktop improved my score by 7-10!


P.S. I tried this several times.
 
Re: Re: spelling

Originally posted by Nermal
*Ahem*

O-P-T-I-M-I-S-E

(for those of you outside the USA) :)
Click to expand...

Yeah, us non-US Mac users do spell things a little differently, but it's a lot simpler, for example:

iTunes Music Store is spelt as:
N-O-T--Y-E-T--A-V-A-I-L-A-B-L-E

iPhoto Printing Service is spelt:
N-O-T--Y-E-T--A-V-A-I-L-A-B-L-E

Most Sherlock channels are spelt:
N-O-T--Y-E-T--A-V-A-I-L-A-B-L-E

See? Simple to remember! :D
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back